A NORMATIVE DATABASE OF

FUNCTIONAL (SHOULDER) TESTS WITHIN

HEALTHY MALE OVERHEAD ATHLETES

Dimitri Audenaert

Jeroen Baele

Joren Christiaens

Supervisors: Prof. Dr. Ann Cools, PhD-student Dorien Borms

A dissertation submitted to Ghent University in partial fulfilment of the requirements for the

degree of Master

Academic year: 2016 – 2017

A NORMATIVE DATABASE OF

FUNCTIONAL (SHOULDER) TESTS WITHIN

HEALTHY MALE OVERHEAD ATHLETES

Dimitri Audenaert

Jeroen Baele

Joren Christiaens

Supervisors: Prof. Dr. Ann Cools, PhD-student Dorien Borms

A dissertation submitted to Ghent University in partial fulfilment of the requirements for the

degree of Master

Academic year: 2016 – 2017

4

Acknowledgements After an intensive academic year, it has finally come so far. By writing our acknowledgements we

finalize our dissertation. Many people have had their share in accomplishing this thesis. Now we

would like this opportunity to express our sincere gratitude for their support and effort.

Firstly, we would like to thank our promotor, Prof. Dr. Ann Cools and co-promotor, PhD-student

Dorien Borms for their excellent guidance during this project. Dorien Borms was always available

to provide us with her expertise and to give us feedback on our work. A special thanks to the

Department of Rehabilitation Sciences and Physiotherapy for letting us use their materials.

Furthermore, our gratitude goes out to all the sports clubs and sports services for letting us use

their facilities and accommodations, also we highly appreciate the motivation of the participating

sports athletes.

Besides, we would like to thank our parents for the financial means that they had to endow. Lots of

transfers had to be made to execute our tests so our parents had to lend us a car.

At last, sincere thanks to our friends and partners for their help and involvement.

Dimitri Audenaert

Jeroen Baele

Joren Christiaens

Ghent, 11 May 2017

5

Contents Acknowledgements ............................................................................................................................ 4

List of figures an tables ....................................................................................................................... 6

List of abbreviations ........................................................................................................................... 7

Abstract (EN) ...................................................................................................................................... 8

Abstract (NL) ....................................................................................................................................... 9

Introduction ...................................................................................................................................... 10

Methods ........................................................................................................................................... 12

Results .............................................................................................................................................. 19

Discussion ......................................................................................................................................... 27

References ........................................................................................................................................ 32

Abstract (NL) - lekentaal ................................................................................................................... 35

Proof of submission ethical commission .......................................................................................... 37

Signed document of laboratory agreements ................................................................................... 49

Signed document of laboratory confidentiality ............................................................................... 50

Attachments ..................................................................................................................................... 53

6

List of figures an tables Figure 1. YBT-UQ with reach directions

Figure 2. YBT-UQ medial reach direction

Figure 3. YBT-LQ reach directions for right dominant limb

Figure 4. Heel stays on the box while reaching

Figure 5. YBT-LQ posterolateral reach direction

Figure 6. CKCUEST initial push-up position

Figure 7. SMBT starting position

Figure 8. Isometric internal rotation

Figure 9. Isometric external rotation

Table 1. Anthropometric data of the study participants

Table 2. Descriptive analysis of the results of the normative and composite scores of the

YBT-UQ

Table 3. Descriptive analysis of the results of the normative and composite scores of the

YBT-LQ

Table 4. Descriptive analysis of the results of the normative scores of the SMBT and mean

scores of CKCUEST

Table 5. Descriptive analysis of the results of the normative and ratio values of the HHD

Table 6. ICC, SD, SEM and MDC of the included tests

Table 7. Results from the GLM ANOVA for repeated measures and post hoc tests for all

variables

7

List of abbreviations ANT Anterior

CKCUEST Closed Kinetic Chain Upper Extremity Stability Test

ER/IR ratio External rotation/internal rotation ratio

GLM-ANOVA General linear model Analysis of Variance

HHD Hand-Held Dynamometer

ICC Intraclass correlation coefficient

IL Inferolateral

LLL Lower limb length

MED Medial

MDC Minimal detectable change

NORM Normalized

PL Posterolateral

PM Posteromedial

SD Standard deviation

SEM Standard error of measurement

SIAS Spina Iliaca Anterior Superior

SL Superolateral

SMBT Seated Medicine Ball Throw

ULL Upper limb length

YBT-LQ Y-Balance Test Lower Quarter

YBT-UQ Y-Balance Test Upper Quarter

8

Abstract (EN)

Background Almost no normative data was available in the literature for the Y-Balance Test Upper Quarter (YBT-

UQ), Y-Balance Test Lower Quarter (YBT-LQ), Closed Kinetic Chain Upper Extremity Stability Test

(CKCUEST), Seated Medicine Ball Throw (SMBT) and the isometric strength test of internal and

external rotation with the Hand-Held Dynamometer (HHD). Normative data can be used for return

to sport criteria and injury prevention.

Objectives This cross-sectional study was to conduct science-based guidelines for return to play concerning

the included tests within healthy overhead athletes. Therefore, the purpose of the study was to

provide a normative database for these tests in a sample of overhead athletes to discuss age, sports

and side differences.

Study Design Cross-sectional study

Method A total of 106 male healthy overhead athletes between 18 and 50 years old (36 volleyball players,

33 tennis players, 37 handball players; age = 25.3 ± 6.6; height = 183.3 ± 6.5; weight = 79.3 ± 10.1)

were included in this study. Functional upper limb performance was assessed using YBT-UQ, SMBT

and CKCUEST with addition of YBT-LQ for the lower limb and a HHD was used to measure isometric

internal and external shoulder strength. Pearson correlation coefficients and coefficients of

determination between the included tests were determined. Analysis of variance (ANOVA, linear

mixed models) for repeated measures were used to analyze differences for the included tests

between arm dominance, age categories and sports disciplines.

Results Normative data showed differences for the sports disciplines, age categories and arm dominance.

No strong correlations were found between the included tests. No relevant three-way interactions

were available, only a few two-way interactions and main effects were present.

Conclusion This study offers a normative database on the included tests. There are differences present

between arm dominance, age categories and sports disciplines. Because no strong correlations

were found between the included tests, it can be assumed that these tests are not measuring the

same aspect of shoulder functionality. Therefore, it can be useful to do all these tests before return

to play and do not take conclusions on one test. Further research is necessary to conduct science-

based guidelines or injury prevention or return to play concerning the included shoulder tests.

9

Keywords Normative database, Y-Balance Test, Closed Kinetic Chain Upper Extremity Test, Seated Medicine

Ball Throw, Hand-Held Dynamometer

Abstract (NL)

Achtergrond In de literatuur zijn bijna geen normatieve data weergegeven voor testen zoals de Y-Balance Test

Upper Quarter (YBT-UQ), Y-Balance Test Lower Quarter (YBT-LQ), Closed Kinetic Chain Upper

Extremity Stability Test (CKCUEST), Seated Medicine Ball Throw (SMBT) en de isometrische

spierkrachttest van de interne en externe schouderrotatoren met de Hand-Held Dynamometer

(HHD). Normatieve data kan gebruikt worden voor return to play criteria en blessurepreventie.

Doelstellingen De doelstelling van deze cross-sectionele studie was om wetenschappelijke richtlijnen voor return

to play op te stellen met betrekking tot de geïncludeerde testen binnen een populatie van gezonde

bovenhandse sporters. De opzet van deze studie was dus om normatieve data te verschaffen voor

deze testen in een steekproef van bovenhandse sporters. Zo kan men de verschillen tussen leeftijd,

sport en zijde bediscussiëren.

Onderzoeksdesign Cross-sectioneel onderzoek

Methode Een totaal van 106 bovenhandse sporters met een leeftijd van 18 tot 50 jaar oud (36

volleybalspelers, 33 tennisspelers, 37 handbalspelers; leeftijd = 25.3 ± 6.6; lengte = 183.3 ± 6.5;

gewicht = 79.3 ± 10.1) werden in deze studie geïncludeerd. De prestatie van het bovenste lidmaat

werd geëvalueerd aan de hand van YBT-UQ, SMBT en CKCUEST met aansluitend de YBT-LQ voor het

onderste lidmaat en de HHD om de isometrische kracht van de interne en externe schouderrotatie

te bepalen. De Pearson correlatie coëfficiënten en determinatiecoëfficiënten tussen de

geïncludeerde testen werden bepaald. Analyse van variantie (ANOVA, linear mixed models) voor

herhaalde metingen werden gebruikt om verschillen te analyseren voor de geïncludeerde testen

qua armdominantie, leeftijdsgroepen en sportdisciplines.

Resultaten In de normatieve data zijn verschillen aantoonbaar voor sport, leeftijd en armdominantie. Geen

sterke correlaties zijn gevonden tussen de functionele testen en de isometrische krachtsvariabelen.

Geen relevante driewegsinteractie was beschikbaar, slechts enkele tweewegsinteracties en

hoofdeffecten waren aanwezig.

10

Conclusies Deze studie biedt normgegevens aan voor de geïncludeerde testen. Er zijn verschillen aantoonbaar

tussen armdominantie, leeftijdsgroepen en de sportdisciplines. Omdat er geen sterke correlaties

zijn aangetoond tussen de functionele testen en de isometrische krachtsvariabelen, kan er gesteld

worden dat deze testen elk een ander aspect van de schouderfunctionaliteit meten. Het kan dus

aangewezen zijn om al deze testen uit te voeren alvorens een conclusie te trekken omtrent return

to play in plaats van zich te baseren op slechts één test. Verder onderzoek is noodzakelijk om

wetenschappelijke richtlijnen op te kunnen stellen voor blessurepreventie of return to play criteria

met betrekking tot de geïncludeerde testen.

Kernwoorden Normatieve database, Y-Balance Test, Closed Kinetic Chain Upper Extremity Test, Seated Medicine

Ball Throw, Hand-Held Dynamometer

Introduction The shoulder is likely to be the most used joint in our daily activities and therefore susceptible to

develop several injuries. Because of a large range of motion a good stability is necessary to avoid

excessive motions which can cause injuries due to chronic stress [1, 2]. Anthropometric measures,

player position, skill level, amount of training time, flexibility, asymmetry, prior injury, playing

surface and shoe type are known risk factors [3-8]. Overhead throwing is an extreme movement

because of the high speeds and use of an excessive range of motion like 150 to 210° of external

rotation in the shoulder joint [9, 10]. A good functionality of the shoulder requires an optimal

coordination of the scapula, stability and kinetic chain function. The integrity of the rotator cuff and

capsular structures must be intact in order to serve a stable center of rotation during the throwing

motion [11]. Due to susceptibility of the shoulder joint it is most likely that the shoulder injuries

occur in overhead sport activities. Common acute disorders are traumatic luxation,

acromioclavicular joint disruption, traumatic tendon ruptures, labral lesions, cartilage defects and

fractures and furthermore chronic disorders like bursitis, tendinitis, rotator cuff tears and shoulder

impingement [12]. Conservative and post-operative injuries demand an intensive and long-term

rehabilitation. In order to return to play, criteria are necessary to objectify rehabilitation goals and

assess the functionality of the shoulder. Literature states that competitive athletes are more prone

to having a faster come back to sports than recreational athletes [13].

Rationale Functional shoulder tests are included in the rehabilitation because of their qualitative and

quantitative assessment that objectifies a person his task or sport specific movements. Different

11

functional shoulder tests are presented in the literature. The advantages of these functional tests

are that they require little experience, no high budget and no high technology. Moreover, these

tests are easily executed and understood by patient and therapist [14]. According to the systematic

literature overview by Audenaert, Baele & Christiaens [15] a diversity in methods and outcome data

were detected within studies for the different functional shoulder tests. Due to a lack of possibility

for comparison between functional shoulder tests, a standardized protocol was established for this

study.

In the literature studies of Teyhen et al. [16], Butler et al. [17], Butler et al. [18] and Gorman et al.

[19] composite scores of the Y-Balance Test Upper Quarter (YBT-UQ) were presented for the

dominant upper limb between 85.7 and 88.3% of the upper limb length (ULL) and between 87.2

and 88.3%ULL for the non-dominant upper limb. In a study of Ali H Alnahdi et al. [20] composite

scores were described for the Y-Balance Test Lower Quarter (YBT-LQ) being 94.1% of the lower limb

length (LLL) for the dominant leg and 95.3%LLL for the non-dominant leg. Concerning the Closed

Kinetic Chain Upper Extremity Stability Test (CKCUEST), values between 22.00 and 30.41 were

reported by Pontillo et al. [21], Roush et al. [22] and Sciascia et al. [23]. No relevant values were

found regarding the Seated Medicine Ball Throw (SMBT) due to protocol variation in starting

position, weight of the ball and throwing angle. According to the study of Cools et al. [24] there

were available values across all athletes and ages in a supine position with 90° of abduction in the

shoulder and neutral rotation. Mean values for isometric external rotation of the dominant arm

with the Hand-Held Dynamometer (HHD) were 145.2N and 140.3N for the non-dominant arm.

Mean values for isometric internal rotation of the dominant arm were 165.4N and 153.2 for the

non-dominant arm. In contrary to the study of Cools et al. [24], this cross-sectional study executed

the isometric shoulder rotation strength test in 0° abduction and neutral rotation. Overall, there is

little availability of normative data of return to play tests in overhead athletes.

In this cross-sectional study the YBT-UQ, YBT-LQ, CKCUEST, SMBT and isometric shoulder rotation

strength with the HHD were executed. Solely the YBT-LQ is not directly shoulder related though

was implemented to evaluate the kinetic chain. It is most likely that functional performance of the

lower limb is related to the performance of upper limb joints due to the influence of the kinetic

chain [25, 26]. In this study an overhead sports population of male volleyball, handball and tennis

players underwent the included tests mentioned above. The age of the subjects varied between 18

to 50 years old which were classified in three age groups: 18-25y, 26-33y and 34-50y. The

comparison was not defined and data of the tests were considered as outcome.

12

Objectives No universal accepted tests for return to play are available for the upper limb in contrary to the

lower limb where tests as ‘single leg hop test’ could predict injury and provide performance values

[27-29]. Reference values of the included tests are essential for physiotherapists and others to

define the rehabilitation goals. Until now there are no available normative values for the YBT-UQ,

YBT-LQ, CKCUEST, SMBT and isometric shoulder rotation test measured with the HHD in a 0-0

position. What are the reference values of the described included tests? A normative database

could be able to determine return to play and to objectify progress. For example, an athlete who

underperformed in total could be susceptible for shoulder pathology. Furthermore this player could

be assessed by these tests to determine rehabilitation process and ability to return to play. The

purpose of this study is to establish a normative database for the included tests in overhead

athletes.

Methods

Setting Test moments took place in the laboratories of Ghent University Hospital or in the sports

accommodations of the involved clubs. The recruitment started in July of 2016 and persisted during

the testing phase, which began at start of the academic year 2016-2017. Data collection started in

February 2017 while the testing phase was still active. The testing phase was finalized at the end of

March.

Participants Inclusion and exclusion criteria were based on the article of Cools et al. [24]. Included subjects were

between 18 and 50 years old and were minimal 3 h/week active in volleyball, tennis or handball on

a competitive level. Subjects were excluded if they had any history of shoulder sub- or dislocation,

orthopedic surgery, systemic diseases or if they experienced pain with time-loss in sports

participation the past 6 months. Inclusion and exclusion criteria were assessed with a

questionnaire. Sports clubs and individual athletes were contacted by e-mail, mobile phone and in

person. An online survey was composed to facilitate the screening of potential participants. After

screening for eligibility, practical matters as date, time, participants and location were agreed on.

The study sample consisted of 106 male healthy overhead athletes. 10 participants were dominant

on the left side and 96 on the right side. The dominance was determined by asking the preferred

throwing hand. One hundred subjects were ought to be sufficient for a reliable study in a proportion

of at least 33 subjects per sports discipline. Furthermore each sports discipline was initially divided

in four groups regarding age: 18-25y, 26-33y, 34-41y and 42-50y. Due to a lack of sufficient

13

N number, BMI Body Mass Index

participants within the oldest categories, they were merged into one category of age 34-50 years.

A presentation of anthropometric data is summarized in Table 1. Approval for the investigation was

received by the Ethical Committee of Ghent University (2016/0963).

Table 1. Anthropometric data of the study participants

Sport Age Category N Age Height Mass BMI

Volleyball 18-25 19 21.8 ± 2.1 184.4 ± 6.9 79.0 ± 7.8 23.3 ± 2.2

26-33 11 29.2 ± 2.6 186.4 ± 5.9 82.2 ± 9.4 23.7 ± 2.4

34-50 6 35.7 ± 2.0 184.8 ± 6.3 78.9 ± 11.3 23.1 ± 3.2

Total, 18-50 36 26.4 ± 5.8 185.1 ± 6.4 80.0 ± 8.8 23.3 ± 2.3

Tennis 18-25 20 20.2 ± 2.0 181.3 ± 5.2 74.4 ± 10.8 22.6 ± 3.0

26-33 6 27.0 ± 2.6 184.2 ± 5.9 77.2 ± 4.6 22.8 ± 1.5

34-50 7 38.3 ± 5.5 177.8 ± 7.1 80.3 ± 9.1 25.4 ± 2.1

Total, 18-50 33 25.3 ± 7.9 181.1 ± 5.9 76.1 ± 9.7 23.2 ± 2.8

Handball 18-25 24 20.8 ± 2.6 183.8 ± 6.1 78.4 ± 10.7 23.1 ± 2.4

26-33 9 27.6 ± 1.4 183.5 ± 8.6 87.3 ± 12.7 26.1 ± 4.6

34-50 4 37.5 ± 3.7 183.6 ± 7.8 87.0 ± 3.2 25.8 ± 1.4

Total, 18-50 37 24.3 ± 5.9 183.7 ± 6.7 81.5 ± 11.3 24.1 ± 3.2

All sports 18-25 63 20.9 ± 2.2 183.2 ± 6.1 77.3 ± 10.0 23.0 ± 2.5

26-33 26 28.1 ± 2.4 184.9 ± 6.8 82.8 ± 10.3 24.3 ± 3.4

34-50 17 37.2 ± 4.1 181.6 ± 7.3 81.4 ± 9.2 24.7 ± 2.6

Total, 18-50 106 25.3 ± 6.6 183.3 ± 6.5 79.3 ± 10.1 23.6 ± 2.8

Testing procedure Prior to participation, the questionnaire was checked for eligibility. If so, subjects read and signed

the informed consent form. The participant determined the randomization of the tests by taking

five cards with the name of the test on the backside. Anthropometric measurements as weight,

height, ULL and LLL were measured before testing. Weight was measured with the same scale for

all subjects. Measurement of the ULL happened with a tape measure in standing position reaching

from the spinous process of the seventh cervical vertebrae to the most distal point of the middle

finger in 90° shoulder abduction, arm fully extended and palm of the hand facing the ventral side.

LLL was measured starting from the SIAS reaching to the distal point of the medial malleolus after

neutralizing lumbopelvic rotation in supine position. Non-dominant side was primary tested to

dominant side for each test. The tests were performed in sports outfit and barefoot.

14

YBT-UQ Materials required to conduct the YBT-UQ were the Y-

Balance Test Kit™ (Move2Perform, Evansville, IN), duplex

tape and a fitness mat. The duplex tape was necessary to

fixate the Y-Balance Test Kit™ on the ground. A fitness mat

was used to make the participant comfortable during the

rest period. The test was explained and demonstrated by

the researchers. The participant had to stand in a three-

point plank position (figure 1) with the shoulder

perpendicular to the hand and with the feet shoulder-width

apart [16-18, 30]. The tested side was named based on the

hand providing support during the trial. The thumb of the

tested arm was placed as close as possible to the red line

without crossing it. The red line was located on the medial

side of the stance platform perpendicular to the medial

reach direction. The athlete started from this position and reached a reach box successively in the

medial (M), inferolateral (IL) and superolateral (SL) direction (figure 1 and 2). The participant had

to reach the reach box on the inner side with controlled speed and without kicking it away. It was

not allowed to touch the floor or the topside of the reach box. During the test, the body could be

adjusted in order to obtain an optimal reach distance as long as the feet and tested upper limb

remained still. The test was completed when the participant was able to retain stability by coming

back to the starting position. After completion, the reach distance was measured in centimeter up

to half a centimeter. Execution of the test was similar to the studies of Butler et al. [17, 18], Gorman

et al. [19], Teyhen et al. [16] and Westrick et al. [30]. The participant had two practice trials on each

side which had to be performed as good as possible to guarantee an optimal result during the

performance trials [16-18]. Standard motivation was applied during the performance. Feedback

could be given during and after the practice trials to endeavor a perfect executed test. A rest period

of one minute was foreseen after the practice trials. There were three performance trials on each

side with a rest period of 30 seconds [16, 19]. The person was instructed to perform as good as

possible accounting the received feedback of the researchers. The primary variables for the test

contained a mean score for each direction separately for the tested side [16, 19]. To be able to

compare the results between the participants, the mean score of each direction was normalized by

the ULL [16, 19]. Subsequently, a composite score for each limb was calculated as a secondary

Figure 1. YBT-UQ with reach directions

Figure 2. YBT-UQ medial reach direction

15

variable in percentage by an addition of the average reach values of the three directions, divided

by three times the ULL [30].

YBT-LQ Materials required to conduct the YBT-LQ were the Y-

Balance Test Kit™ (Move2Perform, Evansville, IN) and

duplex tape. The duplex tape was necessary to fixate the

Y-Balance Test Kit™ on the ground. The test was explained

and demonstrated by the researchers. The participant had

to adapt a single lower limb stance on the stance platform

box, which was the tested side. The top of the hallux had

to be behind the red line. The red line was located on the

anterior side of the immovable central box, perpendicular

to the anterior reach direction. In contrast to the YBT-UQ, the YBT-LQ had to be completed for the

non-dominant and dominant side in one direction before testing the next direction instead of all

directions in one trial. Reaching directions were respectively anterior (A), posteromedial (PM) and

posterolateral (PL) (figure 3, 5). The participant had to reach the reach box on the inner side with

controlled speed and without kicking it away. It was not allowed to touch the floor or the topside

of the reach box. During the test, the body could be

adjusted in order to obtain an optimal reach distance as

long as the heel stayed on the stance platform (figure 4).

The test was completed when the participant was able to

retain stability by coming back to the starting position. After

completion, the reach distance was measured in centimeter

up to half a centimeter. Execution of the test was similar to

the study of Kang et al [31]. The participant had two

practice trials on each side for a specific direction. This was

repeated for every direction before starting the

performance trials. Practice trials, motivation, feedback,

rest periods between practice and performance trials were

applied identically as the YBT-UQ except no rest period was

applied between performance trials. The participants

executed three performance trials [16, 20, 31]. The primary

variables for the test contained a mean score for each direction separately for the tested side [16,

31]. To be able to compare the results between the participants, the mean score of each direction

Figure 3. YBT-LQ reach directions for right dominant limb

Figure 4. Heel stays on the box while reaching

Figure 5. YBT-LQ posterolateral reach direction

16

was normalized by the LLL [16, 20, 31]. Subsequently, a composite score for each limb was

calculated as a secondary variable in percentage by an addition of the average reach values of the

three directions, divided by three times the LLL [20].

CKCUEST Materials required to conduct the CKCUEST were two

strokes of white tape of 3.8cm, a fitness mat and a

stopwatch. There had to be a distance of 91.4cm between

the two strokes of white tape [14, 21-23, 32-34]. The test

was explained and demonstrated by the researchers [34,

35]. An initial push-up position (figure 6) with the middle

fingers on the tape strokes, feet placed shoulder width

apart and the shoulders perpendicular to the hands was

assumed. A straight back had to be maintained during the test [14, 21-23, 32-36]. A fitness mat

could be placed underneath the knees for comfort. The number of the touches was counted out

loud and was considered as a standard motivation during the performance. Feedback could be

given during and after the familiarization trial to endeavor a perfect executed test. By the sign of

the researcher, the subject brought his dominant hand to his non-dominant hand that was

stabilizing at the moment, followed by replacing his dominant hand to the starting position.

Subsequently the non-dominant hand acted in an identical way. This alternating technique was

continued for 15 seconds wherein the participant tried to accomplish as many touches as possible

[14, 21-23, 32-36]. One researcher counted the number of touches and another researcher took

care of the timing [33, 35]. When one researcher said ‘stop’, both researchers focused on the last

real touch to decide whether it was in time or not. Both researchers were aware of possible errors

such as not touching the hand, not placing the hand on the tape, bending the knees on the ground

and failing to maintain feet on the ground or a straight back. There was one submaximal

familiarization trial and four performance trials. Unable to touch the hand, as it should be, was not

counted as one tap. If one of the other errors of above were present, the test was immediately

ejected and a rest period was given to prepare for a new trial. At the start of the performance trials,

the participants had been told to execute the test as good as possible. A rest period of 45 seconds

was present between the performance trials [22, 36]. A mean score was calculated of the three

performance trials [33, 36]. No normalization was applied on the values.

Figure 6. CKCUEST initial push-up position

17

SMBT Materials required to conduct the SMBT were a 2kg medicine

ball [37, 38], a 10m tape measure, chalk powder and a sweeping

brush. Chalk powder was necessary to leave a print behind on

the floor to determine the throwing distance. The 10m tape

measure had to start from a stable wall where the participant

was able to sit down. The test was explained and demonstrated

by the researchers. The participant had to sit down with their

hips as close to the wall as possible. The back, shoulders and

head had to make full contact with the wall at all times [37, 39,

40]. The legs had to be extended and closed (figure 7). To allow

for different arm length between participants, the ball was

dropped with extended arms and horizontally in front of the

subject when the legs were abducted. The distance between the wall and the most proximal point,

indicated by the chalk powder, was subtracted from the total throwing distance. This distance was

considered as the zero mark. The medicine ball was in both hands with the arms in 90° abduction

and full elbow flexion. After permission of the researcher, the subject had to throw the medicine

ball in a horizontal line as far as possible. Presence of a reviewer within a throw lane of 10m was

not allowed to avoid that the subject would withhold his throw. If the back, shoulders or head lost

contact with the wall or if the medicine ball was thrown in a deviating angle instead of a horizontal

line, the test was invalid and needed to be re-done [39, 40]. There were three practice trials after

instruction and demonstration. It was important that the subject delivered maximum performance

to guarantee an optimal result of the performance trials. The researchers were using standardized

instructions and verbal cues. Feedback could be given during and after the practice trials to

endeavor a perfect executed test. There was a rest period of two minutes after the practice trials.

Afterwards there were four performance trials and a rest period of one minute was given after each

performance trial. After each performance trial, the researcher noted the throwing distance to one

centimeter accurately. This was the distance between the wall and the most proximal point

indicated by the chalk powder. Afterwards, the chalk powder had to be removed by a sweeping

brush to be ready for the next trial. The mean score for the SMBT was calculated by dividing all

adapted throwing distances in cm by the four performance trials [38, 40-48]. Note that the original

throwing distance was adapted by subtracting the zero mark distance to eliminate influence of the

arm length.

Figure 7. SMBT starting position

18

HHD Materials required to conduct this test were MicroFET©

HHD and a treatment table. The test was explained and

the researcher showed the shoulder movements. Before

testing, the subject made a standardized warming-up of

ten times a swinging movement to shoulder abduction-

adduction, ten times a swinging movement to shoulder

anteflexion-retroflexion and ten push-ups against the

wall [24]. After the warming-up, the subject was lying

supine on the treatment table [49]. The researcher gave

information about the internal and external rotators

muscles of the arm by showing an internal and external

rotation movement of the arm. The purpose of this

device was also explained to the subjects. The researcher

took the arm of the subject and made an internal and

external movement as kind of a demo. The position that

had to be maintained during the test was a supine position with the legs extended, feet and hips

on the treatment table without losing full contact, opposite hand under the pelvis to avoid

compensation, elbow against the trunk with an elbow flexion of 90°, dorsum of the hand pointed

laterally and hand and fingers relaxed (figure 8,9). During the test, the force had to be build-up

slowly within the first two seconds to reach the maximal force for the following five seconds [24,

49]. It was important that the participant and the researcher dropped the force and lost contact at

the same time, if not, errors could occur. The researcher held the HHD 2cm under the styloid

process which was marked with a pencil (for internal rotation on the palmar side and for external

rotation on the dorsal side) [24]. The test must be an isometric make test. The hand of the

participant was fixated between the chest of the researcher and the hand that held the HHD. The

elbow was fixated against the trunk of the subject for external rotation [50]. The results were

measured up to 0.1 Newton approximately. This test was repeated three times for each muscle

group of one side with a rest period of 20 seconds between the performance trials. If there was a

difference of 20% between the trials of a muscle group on a specific side, a fourth trial was

necessary. The mean score of the three valid performance trials was calculated for the HHD test.

Normalization was applied by dividing the mean score to the body weight.

Figure 9. Isometric external rotation

Figure 8. Isometric internal rotation

19

Statistical methods Means and standard deviations were calculated for all dependent variables as YBT-UQ (cm)

including medial, inferolateral and superolateral direction, YBT-LQ (cm) including anterior,

posteromedial and posterolateral direction, CKCUEST (number of touches), SMBT (cm) and

isometric shoulder strength test with the HHD (Newton) performed on the dominant and non-

dominant side across the participants. Intraclass correlation coefficients (ICC [3,k]) over the

performance trials of each test were calculated to assess trial-to-trial reliability. Standard error of

the measurement (SEM) and the minimal detectable change (MDC) was calculated to examine

absolute reliability. Since the data was normally distributed, confirmed by the Shapiro-Wilk test,

parametric tests were applied. The Levene’s test showed a normal distribution of variance. Finally

the Mauchly’s test for sphericity validated the use of repeated measures analysis of variance

(ANOVA). Pearson correlation coefficient (r) was used to assess possible relationships between the

variables tests. Correlations (r) were categorized weak (≤0.499), moderate (0.50–0.707) or strong

(≥0.707). A coefficient of determination (R²) was used to explore the amount of variability in the

variables of the tests.

A general linear model for three-way ANOVA for repeated measures was used to analyze

differences between the tests. Side (two levels) was used as within-subject factor and sports

discipline (three levels) and age category (three levels) were used as between-subject factors.

Initially three-way interactions were assessed and in case of no interaction two-way interactions

between the involved variables were examined. If there were still no interactions assessed, main

effects for side, sports discipline, or age category were examined. Alpha was set on 0.05 for the

significance. When performing post hoc analyses a Bonferroni procedure were applied when a

significant difference was found. Statistical analyses were completed by using the IBM SPSS

Statistics, version 21 (IBM Corporation, Armonk, NY, USA).

Results Reference values are divided by sports discipline and age category. Mean normalized scores for the

medial, inferolateral, and superolateral directions and composite scores of the YBT-UQ are provided

in Table 2. The medial reach direction scored higher than the inferolateral reach direction followed

by the superolateral reach direction. Mean normalized scores for the anterior, posteromedial, and

posterolateral directions and composite scores of the YBT-LQ are foreseen in Table 3.

Posteromedial reach direction scored higher than the posterolateral reach direction followed by

the anterior direction. Internal rotation showed higher scores than external rotation for these

reference values. Mean normalized scores for the SMBT and mean scores for the CKCUEST are

20

present in Table 4. Mean normalized scores for the internal and external isometric strength test of

the HDD and ER/IR ratio are summarized in Table 5.

21

Table 2. Descriptive analysis of the results of the normative and composite scores of the YBT-UQ

NORM normalized values, COMP normalized composite scores, ND non-dominant side, D dominant side, MED medial, IL inferolateral, SL superolateral

Table 3. Descriptive analysis of the results of the normative and composite scores of the YBT-LQ Sport Age

Category YBT-LQ ANT ND

NORM YBT-LQ PM ND NORM

YBT-LQ PL ND NORM

YBT-LQ ND COMP

YBT-LQ ANT D NORM

YBT-LQ PM D NORM

YBT-LQ PL D NORM

YBT-LQ D COMP

Volleyball 18-25y 58.78 ± 5.19 110.11 ± 7.37 106.60 ± 6.54 91.83 ± 4.69 58.61 ± 4.67 112.12 ± 8.63 108.42 ± 6.38 93.05 ± 5.35 26-33y 60.26 ± 6.82 106.11 ± 10.91 ±103.68 9.20 90.02 ± 8.11 57.80 ± 5.62 107.68 ± 6.71 108.08 ± 8.52 91.19 ± 6.17 34-50y 61.17 ± 4.47 103.98 ± 9.36 ±97.81 8.35 87.66 ± 6.95 56.61 ± 7.48 100.94 ± 7.97 95.90 ± 7.99 84.48 ± 6.55 Tennis 18-25y 58.86 ± 5.15 105.78 ± 8.04 105.04 ± 8.82 89.89 ± 5.98 58.82 ± 5.73 107.08 ± 7.11 107.76 ± 7.11 91.22 ± 5.71 26-33y 61.08 ± 6.68 109.23 ± 8.84 107.06 ± 4.15 92.46 ± 5.10 61.21 ± 8.53 106.14 ± 12.06 106.16 ± 6.29 91.17 ± 6.91 34-50y 55.50 ± 6.23 100.49 ± 9.98 100.90 ± 9.49 85.63 ± 5.77 57.55 ± 6.72 105.09 ± 9.17 102.79 ± 9.31 88.48 ± 6.73 Handball 18-25y 56.80 ± 4.43 104.44 ± 6.38 101.05 ± 6.89 87.53 ± 4.24 55.17 ± 5.08 102.78 ± 8.06 101.76 ± 6.96 86.78 ± 4.65 26-33y 57.98 ± 5.69 106.58 ± 6.10 105.10 ± 11.37 89.88 ± 6.96 57.41 ± 4.58 106.65 ± 10.10 107.31 ± 10.09 90.46 ± 7.08 34-50y 58.43 ± 2.68 108.74 ± 7.39 105.66 ± 2.66 90.94 ± 3.80 57.45 ± 5.85 104.25 ± 9.71 103.96 ± 6.36 88.55 ± 5.00

NORM normalized values, COMP normalized composite scores, ND non-dominant side, D dominant side, ANT anterior, PM posteromedial, PL posterolateral

Sport

Age Category

YBT-UQ MED ND NORM

YBT-UQ IL ND NORM

YBT-UQ SL ND NORM

YBT-UQ ND COMP

YBT-UQ MED D NORM

YBT-UQ IL D NORM

YBT-UQ SL D NORM

YBT-UQ D COMP

Volleyball 18-25y 104.67 ± 5.08 93.3 ± 10.4 73.1 ± 11.5 90.35 ± 7.04 104.21 ± 5.30 93.65 ± 12.93 72.14 ± 9.58 90.00 ± 7.48 26-33y 102.21 ± 2.74 95.17 ± 11.22 72.16 ± 8.91 89.85 ± 5.71 101.12 ± 3.62 95.19 ± 10.20 73.34 ± 7.40 89.89 ± 5.27 34-50y 102.83 ± 4.46 91.71 ± 10.34 66.23 ± 8.50 86.93 ± 4.77 100.39 ± 2.15 90.84 ± 7.80 65.58 ± 11.95 85.60 ± 2.21

Tennis 18-25y 102.88 ± 5.73 93.76 ± 10.25 68.23 ± 7.22 88.27 ± 6.22 103.45 ± 5.87 90.55 ± 10.46 68.95 ± 8.88 87.65 ± 5.62 26-33y 104.73 ± 13.39 102.01 ± 7.03 82.67 ± 15.38 94.75 ± 8.47 105.17 ± 8.16 95.12 ± 10.68 75.64 ± 10.51 91.83 ± 5.73 34-50y 100.29 ± 9.35 91.43 ± 10.74 67.92 ± 11.86 86.55 ± 9.40 103.77 ± 9.72 84.94 ± 11.94 63.86 ± 12.26 84.19 ± 10.23

Handball 18-25y 104.20 ± 5.46 93.72 ± 12.14 71.44 ± 10.81 89.79 ± 7.38 101.96 ± 5.92 91.38 ± 12.63 70.96 ± 11.33 87.82 ± 8.22 26-33y 106.59 ± 4.70 97.22 ± 10.51 67.37 ± 12.90 90.40 ± 7.46 107.21 ± 8.30 91.67 ± 6.83 64.22 ± 10.93 87.70 ± 7.82 34-50y 111.02 ± 7.75 102.60 ± 6.98 77.35 ± 7.95 96.99 ± 7.23 111.47 ± 7.72 103.12 ± 7.87 72.72 ± 9.03 95.77 ± 7.68

22

NORM normalized values, ND non-dominant side, D dominant side, IR internal rotation, ER external rotation

Table 4. Descriptive analysis of the results of the normative scores of the SMBT and mean scores of CKCUEST Sport

Age Category

SMBT MEAN NORM

CKCUEST MEAN

Volleyball 18-25y 311.33 ± 40.66 27.72 ± 3.68 26-33y 298.09 ± 22.99 27.42 ± 2.53 34-50y 306.21 ± 37.58 26.06 ± 1.82

Tennis 18-25y 296.30 ± 38.80 27.63 ±2.32 26-33y 307.08 ± 21.61 28.06 ± 3.73 34-50y 316.82 ± 50.07 25.10 ± 1.18

Handball 18-25y 306.17 ± 36.15 27.87 ± 2.59 26-33y 346.78 ± 46.67 27.00 ± 2.42 34-50y 298.19 ± 55.43 27.42 ± 2.51

NORM normalized values

Table 5. Descriptive analysis of the results of the normative and ratio values of the HHD Sport Age

Category HHD IR

ND NORM HHD ER

ND NORM HHD

ND RATIO HHD IR

D NORM HHD ER

D NORM HHD

D RATIO

Volleyball 18-25y 2.58 ± 0.58 1.91 ± 0.21 76.14 ± 11.83 2.71 ± 0.44 1.94 ± 0.29 72.10 ± 8.84 26-33y 2.33 ± 0.46 1.76 ± 0.30 77.11 ± 15.19 2.61 ± 0.42 1.93 ± 0.25 75.00 ± 10.41 34-50y 2.03 ± 0.32 1.63 ± 0.27 80.67 ± 8.39 2.39 ± 0.39 1.73 ± 0.49 71.42 ± 11.53

Tennis 18-25y 2.58 ± 0.53 1.93 ± 0.32 76.63 ± 13.10 2.94 ± 0.61 2.05 ± 0.30 71.64 ± 13.19 26-33y 2.25 ± 0.30 1.76 ± 0.17 78.67 ± 4.89 2.38 ± 0.58 1.86 ± 0.31 80.33 ± 12.79 34-50y 2.53 ± 0.43 1.89 ± 0.34 75.05 ± 5.95 2.97 ± 0.51 1.93 ± 0.32 65.42 ± 9.67

Handball 18-25y 2.33 ± 0.47 1.79 ± 0.30 78.70 ± 14.67 2.54 ± 0.52 1.85 ± 0.27 74.74 ± 12.97 26-33y 2.49 ± 0.56 2.01 ± 0.42 82.50 ± 17.11 2.77 ± 0.69 2.04 ± 0.63 74.85 ± 21.36 34-50y 2.55 ± 0.42 1.91 ± 0.41 75.43 ± 11.53 2.70 ± 0.76 1.88 ± 0.62 70.89 ± 16.79

23

ND non-dominant side, D dominant side, MED medial, IL

inferolateral, SL superolateral, ANT anterior, PM

posteromedial, PL posterolateral, IR internal rotation, ER

external rotation, ICC intraclass correlation coefficient, SD

standard deviation, SEM standard error of the measurement,

MDC minimal detectable change

The average measures ICC (3,k) for the 5 tests ranged between 0.874 and 0.979, which confirms

the good-to-excellent reliability of these tests. The values for the SD, SEM and MDC are summarized

in Table 6.

Table 6. ICC, SD, SEM and MDC of the included tests

ICC SD SEM MDC

YBT-UQ D MED 0.949 7.137 1.612 4.468 IL 0.963 11.359 2.185 6.056 SL 0.963 10.288 1.979 7.056

ND MED 0.956 7.032 1.475 4.088 IL 0.964 10.540 2.000 5.543 SL 0.959 10.628 2.152 5.965

YBT-LQ D ANT 0.954 5.966 1.280 3.547 PM 0.957 9.047 1.876 5.200 PL 0.874 8.173 2.901 8.042

ND ANT 0.964 6.158 1.168 3.239 PM 0.957 8.623 1.788 4.956 PL 0.947 7.976 1.836 5.090

HHD D IR 0.975 43.294 6.845 18.974 ER 0.966 26.420 4.872 13.504

ND IR 0.979 41.871 6.068 16.819 ER 0.967 23.762 4.317 11.965

SMBT

0.881 39.439 13.605 37.711

CKCUEST

0.927 2.735 0.739 2.048

Correlation Coefficients, Coefficient of Determination and Difference (P Values) between the tests

are shown in the attachments. Generally, independent of the side for the isometric strength tests

with the HHD correlated to the other tests resulted in significant correlations with the composite

score for the YBT-UQ (r range = 0.189-0.228; P < 0.035) and the YBT-LQ (r range = 0.215-0.353; P <

0.028). Only the non-dominant side for both internal and external rotation had a significant weak

correlation with the SMBT (r = 0,233 and 0,198 respectively; P < 0.017 and P < 0.043 respectively).

Significant correlations were weak for the internal and external rotation of the non-dominant side

and the internal rotation of the dominant side to the CKCUEST (r range = 0.243-0.318; P < 0.012).

The non-significant correlations were weak for the isometric strength tests compared to the other

24

tests. Correlations between the isometric strength test variables itself were all significant, including

a strong correlation for the external and internal rotation comparing the dominant and non-

dominant side (r = 0.796; P = 0 and r = 0.781; P = 0) with a coefficient of determination of 0.634 and

0.610 respectively. Also a moderate correlation of 0.634 was found for the ER/IR ratio comparing

the dominant and non-dominant side (P = 0) and a weak to moderate correlation (r range = 0.440-

0.607; P = 0) was found for the other comparisons between the HHD. Weak correlations were found

for the SMBT compared to YBT-UQ, YBT-LQ and CKCUEST, although they were not significant.

Concerning the CKCUEST, only a significant weak correlation was found with the composite score

of the dominant side of the YBT-UQ. The correlation was weak and not significant for the non-

dominant side. The CKCUEST showed a weak correlation with the composite scores of the YBT-LQ

of both sides but were not significant. Only a significant correlation (r range = 0.227-0.309; P < 0.02)

was present for the posteromedial direction of the dominant side and for posteromedial and

posterolateral direction of the non-dominant side. Correlations between other directions were also

weak but not significant. Correlations between the YBT-UQ and the YBT-LQ were all significant

including a weak correlation (r range = 0.335-0.435) between the composite score of the YBT-UQ

compared to the YBT-LQ for the dominant side as well as for the non-dominant side of each test.

At last, a strong correlation was found between the composite score for the dominant and non-

dominant limb of the YBT-UQ and YBT-LQ (r = 0.812; P = 0 and r = 0.815; P = 0 respectively) with a

coefficient of determination of 0.659 and 0.664 respectively.

The results of the ANOVA for repeated measures statistical analysis and the post hoc tests are

summarized in Table 7. The YBT-UQ showed age and sports discipline differences with significantly

higher composite scores for handball players compared to tennis players in the oldest age category.

All age categories of the handball and tennis players scored significantly higher for the non-

dominant side compared to the dominant side. Concerning the handball players, the medial

direction scored significantly higher in the oldest age category compared to the youngest age

category. Within the oldest age category, the handball players scored significantly higher on the

medial direction compared to the other sports. The medial direction was significantly higher on the

dominant side for the handball players compared to the dominant side of volleyball players and

also the non-dominant side of the handball players was higher compared to the non-dominant side

of tennis players. The inferolateral direction scored significantly higher on the non-dominant side

compared to the dominant side in all sports categories in the youngest and the middle age category,

plus this was also the case for the oldest category in tennis players. The superolateral direction

scored significantly higher on the dominant side compared to the non-dominant side for handball

players. The YBT-LQ showed age and sports discipline differences with significantly higher

25

composite scores in the youngest age category of volleyball players when compared to their oldest

age category. Within the youngest age category, the volleyball players scored significantly higher

than the handball players. The anterior direction scored significantly higher on the non-dominant

side compared to the dominant side for all age categories of the volleyball players. The youngest

age category of volleyball players scored significantly higher than the oldest age category for the

posterolateral direction. A three-way interaction was present for the posteromedial direction. The

isometric internal rotation strength testing showed age and sides differences with significantly

higher scores for the dominant side in every age category. There were also sports discipline and

side differences with significantly higher scores for the dominant side in every sports discipline. The

isometric external rotation strength showed age and side differences with significantly higher

scores for the dominant side in the youngest and middle age category. Only in every age category

of volleyball players was their dominant side significantly higher than their non-dominant side. No

interactions or main effects were found for the SMBT and CKCUEST.

26

Table 7. Results from the GLM ANOVA for repeated measures and post hoc tests for all variables YBT-UQ COMP MED NORM IL NORM SL NORM YBT-LQ COMP ANT NORM PM NORM PL NORM

Three-way interaction NS NS NS NS NS NS NS

Side x sport x age category °

Two-way interaction

NA

Side x sport H°, T°: ND > D D°: H > V ND°: H > T

T*:ND > D H°: D > ND NS V*: ND > D NS

Side x age category NS A°, B: ND > D NS NS NS NS

Sport x age category C°: H > T H°: C > A C: H > V°, H > T°

NS NS V°: A > C A: V > H

NS V: A > C

Main Effects NA NA NA NA NA NA NA NA

Side

Sport

Agecategory

HHD IR NORM HHD ER NORM SMBT NORM CKCUEST MEAN

Three-way interaction NS NS NS NS

Side x sport x age category

Two-way interaction

NA NA

Side x sport H, T*, V*: D > ND V°: D > ND

Side x age category A*, B, C*: D > ND A°, B°: D > ND

Sport x Age category NS

Main Effects NA NA NA NA

Side

Sport

Agecategory

NS not significant, NA not applicable, H handball, T tennis, V volleyball, > higher values, NORM normalized values, COMP normalized composite scores,

ND non-dominant side, D dominant side, A [18-25] age category, B [26-33] age category, C [34-50] age category, MED medial, IL inferolateral, SL

superolateral, ANT anterior, PM posteromedial, PL posterolateral, IR internal rotation, ER external rotation

* p < 0.001; p < 0.01; °p < 0.05

27

Discussion

Summary of evidence This study was the first to have examined reference values of the included five tests. The main

outcome of this study were the normative data of the five tests divided by three sports and three

age categories. There was a good-to-excellent trial-to-trial reliability for all these tests. There were

only significant two-way interaction effects and strong interlimb correlations for the YBT-UQ, YBT-

LQ, and isometric internal and external strength test of the HHD. No strong correlations were found

for all the other variables of the included tests.

YBT-UQ Normative values of the two studies of Butler et al. [17, 18] in male college swimmers and high

school baseball players respectively were 100 ± 8.8%ULL and 99.1 ± 8.6%ULL for the medial

direction, 89.8 ± 10.8%ULL and 90.3 ± 11.5%ULL for the inferolateral direction, 74.9 ± 9.7%ULL and

70.4 ± 11.1%ULL for the superolateral direction and 88.3 ± 8.9%ULL and 86.6 ± 8.1%ULL for the

composite score. These reference values are in line with the scores of our study. The medial

direction was defined as the highest mean value followed by the inferolateral direction and the

superolateral direction, which is in line with previous studies [17-19, 30].

Previous studies [17-19, 30] found no significant differences for the specific directions and the

composite scores between the dominant and non-dominant upper limb in respectively male

overhead athletes, baseball players, active adults and college students while diverging results were

present for arm dominance in sports and age categories in our study. Since there were significant

differences present for dominance of arm, the non-injured arm cannot be used as a reference for

the injured arm while Westrick et al. [30] identified the opposite.

Teyhen et al. [16] concluded that age had no influence on the performance, this is in contrast to

our study where the YBT-UQ showed age and sports discipline differences for the medial and

composite score. Note that only a few number of older handball players were tested in comparison

with the number of younger handball players hence an outlier could have more influence on the

total results which could lead to a higher susceptibility in change of results.

In contrast to Westrick et al. [30], which found a significant weak correlation for both limbs between

YBT-UQ and CKCUEST, only a significant weak correlation was found for the dominant side of the

YBT-UQ with the CKCUEST in our study. The lack of correlation between the YBT-UQ and the other

included tests is probably due to the closed chain performance of the YBT-UQ, whereas the other

included tests were performed in an open kinetic chain, except for the CKCUEST. More specifically,

28

it has been shown that shoulder isometric [30] or isokinetic [51] rotation strength is not related to

the YBT-UQ, suggesting that other variables within the kinetic chain or important for YBT-UQ

performance.

YBT-LQ Normative values for the composite score presented in previous studies were 98.3 ± 8.9%LLL for a

military population [16], 103.0 ± 8.0%LLL for high school basketball players [52] and 94.7 ± 7.0%LLL

for young healthy adults [20]. Reference values in our study were lower than in the previous

mentioned studies. This is in line with previous publications which demonstrated differences on

performance on the YBT-LQ by sports, gender and competition level [53]. Because of sports specific

characteristics, differences could be present between the different sports. In this study, no

differentiation was made between competition levels, they only had to be active in a competition.

This can be a possible explanation for diverging values.

The study of Alnahdi et al. [20] concludes that comparison of the injured leg to the non-injured leg

is possible in young healthy adults, except for the posteromedial direction because in this direction

the non-dominant side scored significantly higher compared to the dominant side. In contrary to

our study, a comparison of the injured leg to the non-injured leg is possible for all directions in all

sports, except for the anterior direction for volleyball players because the non-dominant side

scored significantly higher compared to the dominant side in this sports discipline. No specific

reason can be assumed why volleyball players performed better for the anterior direction than

other sports disciplines. Teyhen et al. [16] concluded that age had no influence on the performance,

this is in contrast to our study where the YBT-LQ showed age and sports discipline differences. Due

to a three-way-interaction of the posteromedial direction, a value can be predicted if at least two

out of three variables are known. Teyhen et al. [16] found a significant moderate correlation

between the YBT-LQ and YBT-UQ, this is in contrast to our study where there was a significant weak

correlation.

CKCUEST Normative data of the CKCUEST in previous studies were 22.5 ± 4.3 touches in division I collegiate

football players [21] and 30.41 ± 3.87 touches in collegiate baseball players [22]. In our study, the

mean result lay in between with a range of 25.10-28.06 touches, achieved over all age categories

and sports. A differentiation in population could explain the variability in results.

The CKCUEST required a combination of shoulder stability but foremost a lot of explosive power

which can clarify no high correlation with other tests. The SMBT required more explosive power

than stability, on the other hand, the YBT tests required more stability than explosivity. This could

29

clarify that no high correlations were found between the CKCUEST and the other included tests.

In our study, a weak correlation was present between the CKCUEST and the isometric strength test

while in other studies there was a high correlation between the CKCUEST and the peak torque of

isokinetic internal/external rotation [54]. As already mentioned, difference in type of contraction

(isokinetic vs. isometric) could be the reason for these divergent findings.

Note that more volleyball and tennis players of age category [25-33y] and [34-50y] were tested

compared to the number of tested handball players. Therefore a worse performance was noticed

in general for volleyball and tennis players due to older ages. Also a few handball players (n=4)

participated of the last age category [35-50y], which could lead to a higher susceptibility for change

in results.

SMBT Comparison of results of this study to other studies was difficult because of differences in ball

weights, populations and position of throwing. A mean throwing distance of 347.77 ± 76.79cm was

found in a study of Borms et al. [51] in a population of 14 male and 15 female throwing athletes.

The protocol was the same as in this study, although female throwing athletes were included. A

mean throwing distance of 308.32 ± 39.44cm was measured in our study. A possible explanation

for the higher results in the study of Borms et al. [51] could be the difference in competition level

and sample selection bias.

Recently a new variant of the MBT was published in a study by M. Sayers & S. Bishop [55]. This test

was called the Medicine Ball Push-Press (MBP-P) where subjects had to throw the medicine ball

straight perpendicular in supine position. The advantage of this test was that it is easier for the

subjects to throw in a straight vertical line instead of throwing in a horizontal line. For the SMBT,

subjects were triggered to perform as good as possible and due to this, their throw often deviated

from the horizontal line. When they would have to throw as high as possible for the MBP-P, they

would be obligated to throw in a straight vertical line, otherwise their throw would be lower. The

disadvantages of this test were the dysfunctional position and the prices of the high technology

cameras to determine the power, velocity, acceleration and maximal force.

The SMBT showed weak correlations with the included tests, only the correlations with the

isometric strength test variables were significant. In contrary to our study, a study by Borms et al.

[51] showed that the SMBT was moderately to strongly correlated with isokinetic tests for strength

of the elbow flexors and extensors and shoulder internal and external rotation. An explanation for

these contrary results was likely to be that Borms et al. [51] utilized an isokinetic dynamometer

30

(model 4; Biodex Medical Systems Inc, Shirley, NY) to determine the rotational strength in contrast

to our study where the rotational strength was measured manually with a HHD applying an

isometric make-test.

HHD At present, no representative data was available in the literature which studied isometric internal

rotation and external rotation in a 0° abduction and 0° ER position of the upper limb (0-0 position)

in a supine position.

Mean results of all overhead athletes in the study of Cools et al. [24] for the internal rotation

strength were 2.1N/kg for the dominant and 1.9N/kg for the non-dominant side, external rotation

strength for both dominant and non-dominant side were 1.8N/kg. In our study where a 0-0 position

was applied in contrary to the 90-0 position in Cools et al. [24], the range was 2.38-2.97N/kg for

internal rotation and 1.63-2.05N/kg for external rotation. This sets that the internal rotation rotator

strength could be better in a 0-0 position compared to a 90-0 position. Cools et al. [24] calculated

the ER/IR ratio with 89% for the dominant side and 93% for the non-dominant side. In our study, a

range of 65-80% for the dominant side and 75-83% for the non-dominant side was calculated

between the different sports disciplines. A lower ratio was determined due to the difference in

position. A lower ratio is normal due to the 0-0 position where a norm value is 75% while the norm

value for the 90-0 position is 90-100%.

A moderate correlation was found for the ratios of the dominant side compared to non-dominant

side meaning that in several cases an interlimb difference was present due to a disbalance between

external and internal rotation strength. This is in contrary to the YBT tests and the isometric

strength test where a high correlation was calculated between the dominant and non-dominant

sides. Current studies have shown that the throwing movement initiates greater gains in internal

rotation muscles while external rotation muscle strength remains the same which indicates a

disturbed lower ER/IR ratio [56] likewise in our study where the dominant side had a lower ratio

compared to the non-dominant side.

Limitations First, different overhead sports disciplines were included which could have influenced the results

because of sport specific characteristics. Second, athletes older than 26 years old were difficult to

recruit because of the strict inclusion criteria. Older participants are more susceptible for injuries

the last six months and for history of orthopedic surgery, there were also few older than younger

athletes on a competitive level. Third, the results of the tests are dependent of the intrinsic

motivation of the test person, nevertheless a standardized motivation was applied for every test to

31

maximize the performances. Besides, three investigators were part of this research therefore

variance in administering and supervising during the testing was possible. Therefore, to diminish

the effect of possible bias, the investigators were instructed and trained to apply a standardized

protocol. Also the sitting height (SIAS to ball height) was not measured and taken into account

regarding normalization for SMBT. CKCUEST and SMBT were performed bilaterally with the

consequence that no side-to-side differences could be examined. Furthermore, in contrast to all

other included tests, the CKCUEST was not normalized despite that shoulder width and arm length

had an influence on the number of touches. At last, in a study of Kang et al. [31] were ankle

dorsiflexion and hip flexion indicated as the best predictors of respectively the anterior and

posterior reaches. The predictive values were increased by the addition of counterbalanced trunk

kinematics. This findings were similar to the findings of Hoch et al. [57] in the Star Excursion Balance

Test (SEBT). Measuring dorsiflexion range of motion, as flexibility, could help future studies to

differentiate the amount of influence on stability that is part of the final functionality of the lower

limb.

Conclusion This study offers a normative database on functional tests and isometric internal and external

shoulder rotation strength in three age categories and three sports for male overhead athletes.

There are differences present between the three sports and age categories. Because no strong

correlations were found between the functional tests and the isometric strength variables, it can

be assumed that these tests are not measuring the same aspect of shoulder functionality.

Therefore, it can be useful to do all these tests before return to play and do not take conclusions

on one test. Further research is necessary to conduct science-based guidelines or injury prevention

or return to play concerning the included shoulder tests.

32

References 1. Stevenson, J.H. and T. Troijan, Evaluation of shoulder pain. The journal of family practice,

2002. 2. Gerber, C. and A. Sebesta, Impingement of the deep surface of the subscapularis tendon

and the reflection pulley on the anterosuperior glenoid rim: a preliminary report. J Shoulder Elbow Surg, 2000. 9(6): p. 483-90.

3. Faude, O., et al., Risk factors for injuries in elite female soccer players. Br J Sports Med, 2006. 40(9): p. 785-90.

4. Giza, E., et al., Injuries in women's professional soccer. Br J Sports Med, 2005. 39(4): p. 212-6; discussion 212-6.

5. Heidt Jr., R.S., et al., Avoidance of Soccer Injuries with Preseason Conditioning. The American journal of sports medicine, 2000. 28.

6. McGuine, T., Sports Injuries in High School Athletes: A Review of Injury-Risk and Injury-Prevention Research. Clin J Sport Med, 2006.

7. Murphy, D.F., D.A.J. Conolly, and B.D. Beynnon, Risk factors for lower extremity injury: a review of the literature. Br J Sports Med, 2003.

8. Olsen, L., Strategies for prevention of soccer related injuries: a systematic review. British Journal of Sports Medicine, 2004. 38(1): p. 89-94.

9. Feltner, M. and J. Dapena, Dynamics of the Shoulder and Elbow Joints of the Throwing Arm During a Baseball Pitch. International journal of sport biomechanics, 1986.

10. Fleisig, G.S., et al., Kinetics of baseball pitching with implications about injury mechanisms. Am J Sports Med, 1995. 23(2): p. 233-9.

11. van der Hoeven, H. and W.B. Kibler, Shoulder injuries in tennis players. Br J Sports Med, 2006. 40(5): p. 435-40; discussion 440.

12. Doyscher, R., et al., [Acute and overuse injuries of the shoulder in sports]. Orthopade, 2014. 43(3): p. 202-8.

13. Wiese-bjornstal, D.M., et al., An integrated model of response to sport injury: Psychological and sociological dynamics. Journal of Applied Sport Psychology, 1998. 10(1): p. 46-69.

14. Tucci, H.T., J. Martins, and P.M. Camarini, Closed Kinetic Chain Upper Extremity Stability test (CKCUES test): a reliability study in persons with and without shoulder impingement syndrome. BMC Musculoskeletal Disorders, 2014.

15. Cools, A.M., et al., Method and outcome data of the functional shoulder tests YBT-UQ, CKCUEST, SMBT and OAHT and influence on the standardization: a systematic review. 2016.

16. Teyhen, D.S., et al., Normative data and the influence of age and gender on power, balance, flexibility, and functional movement in healthy service members. Mil Med, 2014. 179(4): p. 413-20.

17. Butler, R., et al., Sex differences in dynamic closed kinetic chain upper quarter function in collegiate swimmers. J Athl Train, 2014. 49(4): p. 442-6.

18. Butler, R.J., et al., Bilateral differences in the upper quarter function of high school aged baseball and softball players. The InterThe International Journal of Sports Physical Therapynational Journal of Sports Physical Therapy, 2014.

19. Gorman, P.P., et al., Upper Quarter Y Balance Test: reliability and performance comparison between genders in active adults. J Strength Cond Res, 2012. 26(11): p. 3043-8.

20. Alnahdi, A.H., et al., Reference values for the Y Balance Test and the lower extremity functional scale in young healthy adults. J. of Physical Therapy Science, 2015.

21. Pontillo, M., B.A. Spinelli, and B.J. Sennett, Prediction of in-season shoulder injury from preseason testing in division I collegiate football players. Sports Health, 2014. 6(6): p. 497-503.

33

22. Roush, J.R., J. Kitamura, and M. Chad, Reference Values for the Closed Kinetic Chain Upper Extremity Stability Test (CKCUEST) for Collegiate Baseball Players. North American journal of sports physical therapy, 2007.

23. Sciascia, A. and T. Uhl, Reliabillity of strength and performance testing measures and their ability to differentiate persons with and without shoulder symptoms. The International Journal of Sports Physical Therapy, 2015.

24. Cools, A.M., et al., Eccentric and isometric shoulder rotator cuff strength testing using a hand-held dynamometer: reference values for overhead athletes. Knee Surg Sports Traumatol Arthrosc, 2016. 24(12): p. 3838-3847.

25. Burkhart, S.S., C.D. Morgan, and W. Ben Kibler, The disabled throwing shoulder: spectrum of pathology part III: the SICK scapula, scapular dyskinesis, the kinetic chain, and rehabilitation. Arthroscopy: The Journal of Arthroscopic & Related Surgery, 2003. 19(6): p. 641-661.

26. Chu, S.K., et al., The Kinetic Chain Revisited: New Concepts on Throwing Mechanics and Injury. PM R, 2016. 8(3 Suppl): p. S69-77.

27. Augustsson, J., et al., Single-leg hop testing following fatiguing exercise: reliability and biomechanical analysis. Scand J Med Sci Sports, 2006. 16(2): p. 111-20.

28. Brumitt, J., et al., Lower extremity functional tests and risk of injury in division III collegiate athletes. The International Journal of Sports Physical Therapy, 2013.

29. Earl, J.E., S.K. Monteiro, and K.R. Snyder, Differences in lower extremity kinematics between a bilateral drop-vertical jump and a single-leg step-down. J Orthop Sports Phys Ther, 2007. 37(5): p. 245-52.

30. Westrick, R.B., et al., Exploration of the Y-Balance Test for assessment of upper quarter closed kinetic chain performance. The International Journal of Sports Physical Therapy, 2012.

31. Kang, M.H., et al., Relationship Between the Kinematics of the Trunk and Lower Extremity and Performance on the Y-Balance Test. PM R, 2015. 7(11): p. 1152-8.

32. Lust, K.R., M.A. Sandrey, and S.M. Bulger, The Effects of 6-Week Training Programs on Throwing Accuracy, Proprioception, and Core Endurance in Baseball. Journal of Sport Rehabilitation, 2009.

33. Lee, D. and L.J. Kim, Reliability and validity of the closed kinetic chain upper extremity stability test. Journal of Physical Therapy Science, 2015.

34. Tucci, H.T., et al., Biomechanical Analysis of the Closed Kinetic Chain Upper-Extremity Stability Test. J Sport Rehabil, 2015. 26(1): p. 42-50.

35. Negrete, R.J., et al., Can upper extremity functional tests predict the softball throw for distance: a predictive validity investigation. International journal of sports physical therapy, 2011.

36. Schulte-Edelman, J.A., et al., The effects of plyometric training of the posterior shoulder and elbow. Journal of Strength and Conditioning Research, 2005.

37. Jones, M.T., Progressive-overload whole-body vibration training as part of periodized, off-season strength training in trained women athletes. Journal of Strength and Conditioning Research, 2014.

38. Talukdar, K., et al., The role of rotattional mobility and power on throwing velocity. Journal of Strength and Conditioning Research, 2015.

39. Sreckovic, S., et al., Evaluation of force-velocity and power-velocity relationship of arm muscles. Eur J Appl Physiol, 2015. 115(8): p. 1779-87.

40. van den Tillaar, R. and M.C. Marques, Reliabillity of seated and standing throwing velocity using differently weighted medicine balls. Journal of Strength and Conditioning Research, 2013.

41. Harris, C., et al., The seated medicine ball throw as a test of upper body power in older adults. 2011.

34

42. Lester, M.E., et al., Effect of specific short-term physical training on fitness measures in conditioned men. Journal of Strength and Conditioning Research, 2014.

43. Santos, J.A.M.E. and A.A.S.M. Janeira, The effects of plyometric training followed by detraining and reduced training periods on explosive strength in adolescent male basketball players. Journal of Strength and Conditioning Research, 2011.

44. Santos, J.A.M.E. and A.A.S.M. Janeira, Effects of reduced training and detraining on upper and lower body explosive strength in adolescent male basketball players. Journal of Strength and Conditioning Research, 2009.

45. Santos, J.A.M.E. and A.A.S.M. Janeira, The effects of resistance training on explosive strength indicators in adolescent basketball players. Journal of Strength and Conditioning Research, 2012.

46. Sharp, M.A., et al., Physical fitness and body composition after a 9-month deployment to Afghanistan. Med Sci Sports Exerc, 2008. 40(9): p. 1687-92.

47. Sporis, G., et al., Effects of a training program for special operations battalion on soldiers' fitness characteristics. Journal of Strength and Conditioning Research, 2012.

48. Stockbrugger, B.A. and R.G. Haennel, Contributing factors to performance of medicine ball explosive power test: a comparison between jump and nonjump athletes. Journal of Strength and Conditioning Research, 2003.

49. Andrews, A.W., et al., Normative Values for Isometric Muscle Force Measurements Obtained With Hand-held Dynamometers. Physical Therapy, 1996. 76.

50. Beshay, N., P.H. Lam, and G.A.C. Murrell, Assessing the reliability of shoulder strength measurement: hand-held versus fixed dynamometry. Shoulder & Elbow, 2011. 3(4): p. 244-251.

51. Borms, D., A. Maenhout, and A.M. Cools, Upper Quadrant Field Tests and Isokinetic Upper Limb Strength in Overhead Athletes. J Athl Train, 2016. 51(10): p. 789-796.

52. Plisky, P.J., et al., Star Excursion Balance Test as a Predictor of Lower Extremity Injury in High School Basketball Players. Journal of Orthopaedic & Sports Physical Therapy, 2006.

53. Lehr, M.E., et al., Field-expedient screening and injury risk algorithm categories as predictors of noncontact lower extremity injury. Scand J Med Sci Sports, 2013. 23(4): p. e225-32.

54. Alizadehkhaiyat, O., et al., Shoulder muscle activation and fatigue during a controlled forceful hand grip task. J Electromyogr Kinesiol, 2011. 21(3): p. 478-82.

55. Sayers, M.G. and S. Bishop, Reliability of a New Medicine Ball Throw Power Test. J Appl Biomech, 2017: p. 1-15.

56. Hurd, W.J. and K.R. Kaufman, Glenohumeral rotational motion and strength and baseball pitching biomechanics. J Athl Train, 2012. 47(3): p. 247-56.

57. Hoch, M.C., S.L. Gaven, and J.T. Weinhandl, Kinematic predictors of Star Excursion Balance Test performance in individuals with chronic ankle instability Clinical biomechanics, 2016.

35

Abstract (NL) - lekentaal

Achtergrond In de literatuur zijn bijna geen normgegevens weergegeven voor testen zoals de Y-Balance Test

Upper Quarter (YBT-UQ), Y-Balance Test Lower Quarter (YBT-LQ), Closed Kinetic Chain Upper

Extremity Stability Test (CKCUEST), Seated Medicine Ball Throw (SMBT) en de spierkrachttesten

van inwendige en uitwendige schouderrotatoren zonder verandering in spierlengte met de Hand-

Held Dynamometer (HHD). Normwaarden kunnen gebruikt worden voor sporthervattingscriteria

en blessurepreventie.

Doelstelling De doelstelling van deze cross-sectionele studie was om wetenschappelijke richtlijnen voor

sporthervatting op te stellen met betrekking tot de geïncludeerde testen binnen een populatie van

gezonde bovenhandse sporters. De opzet van deze studie was dus om normwaarden te verschaffen

voor deze testen in een steekproef van bovenhandse sporters. Zo kan men de verschillen tussen

leeftijd, sport en zijde bediscussiëren.

Onderzoeksdesign Cross-sectioneel onderzoek. Hierbij wordt ieder individu in een groep eenmaal en op hetzelfde

tijdstip geobserveerd of gemeten.

Methode Een totaal van 106 bovenhandse sporters met een leeftijd van 18 tot 50 jaar oud (36 volleybal

spelers, 33 tennis spelers, 37 handbal spelers; leeftijd = 25.3 ± 6.6; lengte = 183.3 ± 6.5cm; gewicht

= 79.3 ± 10.1kg) werden in deze studie geïncludeerd. De prestatie van het bovenste lidmaat werd

geëvalueerd aan de hand van YBT-UQ, SMBT en CKCUEST met aansluitend de YBT-LQ voor het

onderste lidmaat en de HHD om de kracht zonder spierlengteverandering van de inwendige en

uitwendige schouderrotatie te bepalen. De Pearson correlatie coëfficiënten en

determinatiecoëfficiënten tussen de geïncludeerde testen werden bepaald. Analyse van variantie

(ANOVA, linear mixed models) voor herhaalde metingen werd gebruikt om verschillen te analyseren

voor de geïncludeerde testen qua armdominantie, leeftijdsgroepen en sportdisciplines.

Resultaten In de normatieve data zijn verschillen aantoonbaar voor sport, leeftijd en zijde. Geen sterke

correlaties zijn gevonden tussen de functionele testen en de isometrische krachtsvariabelen. Geen

relevante driewegsinteractie was beschikbaar, slechts enkele tweewegsinteracties en

hoofdeffecten waren aanwezig.

Conclusies Deze studie biedt normatieve data aan voor functionele testen en isometrische interne en externe

36

schouderrotatiekracht. Er zijn verschillen aantoonbaar tussen dominante en niet-dominante zijde,

leeftijdscategoriën en de sportdisciplines. Omdat er geen sterke correlaties zijn aangetoond tussen

de functionele testen en de isometrische krachtsvariabelen, kan er gesteld worden dat deze testen

elk een ander aspect van de schouderfunctionaliteit meten. Het kan dus aangewezen zijn om al

deze testen uit te voeren alvorens een conclusie te trekken omtrent sporthervattingscriteria in

plaats van zich te baseren op slechts één test. Verder onderzoek is noodzakelijk om

wetenschappelijke richtlijnen te kunnen opstellen voor blessurepreventie of

sporthervattingscriteria met betrekking tot de geïncludeerde testen.

Kernwoorden Normatieve database, Y-Balance Test, Closed Kinetic Chain Upper Extremity Test, Seated Medicine

Ball Throw, Hand-Held Dynamometer

37

Proof of submission ethical commission

38

39

40

41

42

43

44

45

46

47

48

49

Signed document of laboratory agreements

50

Signed document of laboratory confidentiality

51

52

53

COMP normalized composite scores, ND non-dominant side, D dominant side, MED medial, IL

inferolateral, PM posteromedial, PL posterolateral, IR internal rotation, ER external rotation, r

correlation, R2 Coefficient of Determination

Coefficient

Attachments Table X. Correlation Coefficients, Coefficient of Determination and Difference (P Values) between

the functional tests

Variable 1 Variable 2 r R² P Value

HHD ER ND

HHD ER ND

HHD ER D

HHD IR ND

HHD IR ND

HHD IR ND

HHD ER ND

HHD ER D

HHD ER D

HHD IR ND

HHD IR ND

HHD IR D

HHD ER ND

HHD IR ND

HHD ER ND

HHD IR ND

HHD IR D

CKCUEST

CKCUEST

CKCUEST

CKCUEST

YBT-UQ COMP ND

YBT-UQ COMP ND

YBT-UQ COMP D

YBT-UQ COMP D

YBT-UQ COMP ND

YBT-LQ COMP ND

HHD ER ND

HHD IR ND

HHD RATIO D

HHD ER ND

HHD ER ND

HHD ER D

HHD ER D

YBT-UQ COMP ND

YBT-UQ COMP D

YBT-UQ COMP D

YBT-UQ MED ND

YBT-UQ MED D

YBT-UQ IL D

YBT-LQ COMP ND

YBT-LQ COMP ND

YBT-LQ COMP D

YBT-LQ COMP ND

YBT-LQ COMP D

YBT-LQ COMP D

SMBT

SMBT

CKCUEST

CKCUEST

CKCUEST

YBT-UQ COMP D

YBT-LQ PM ND

YBT-LQ PL ND

YBT-LQ PM D

YBT-LQ COMP ND

YBT-LQ COMP D

YBT-LQ COMP ND

YBT-LQ COMP D

YBT-UQ COMP D

YBT-LQ COMP D

HHD ER D

HHD IR D

HHD RATIO ND

HHD IR ND

HHD IR D

HHD IR ND

HHD IR D

0.217

0.228

0.222

0.221

0.208

0.205

0.264

0.271

0.305

0.215

0.353

0.307

0.233

0.198

0.243

0.318

0.281

0.206

0.292

0.227

0.309

0.346

0.335

0.424

0.435

0.812

0.815

0.796

0.781

0.634

0.569

0.607

0.440

0.594

0.047

0.052

0.049

0.049

0.043

0.042

0.070

0.073

0.093

0.046

0.125

0.094

0.054

0.039

0.059

0.101

0.079

0.042

0.085

0.052

0.095

0.120

0.112

0.180

0.189

0.659

0.664

0.634

0.610

0.402

0.324

0.368

0.194

0.353

0.026

0.020

0.023

0.023

0.034

0.035

0.007

0.005

0.002

0.028

0

0.001

0.017

0.043

0.012

0.001

0.004

0.036

0.002

0.020

0.001

0

0.001

0

0

0

0

0

0

0

0

0

0

0